Simian immunodeficiency virus-based lentivirus vector for retinal gene transfer: a preclinical safety study in adult rats

Article metrics


Although lentivirus vectors hold promise for ocular gene therapy, they also have potential safety issues, particularly in the case of the current human immunodeficiency virus-based vectors. We recently developed a novel lentivirus vector derived from the nonpathogenic simian immunodeficiency virus from African green monkeys (SIVagm) to minimize these potentials. In this preclinical study, we evaluated whether SIV vector could be efficiently and safely applicable to retinal gene transfer by assessing the transgene expression, retinal function and histology over a 1-year period following subretinal injection in adult rats. The functional assessment via electroretinogram after both titers of SIV-lacZ (2.5 × 107 or 2.5 × 108 transducing units/ml) injection revealed both the dark and light adaptations to soon be impaired, in a dose-dependent manner, after a buffer injection as well, and all of them recovered to the control range by day 30. In both titers tested, the retinas demonstrated a frequent transgene expression mainly in the retinal pigment epithelium; however, the other retinal cells rarely expressed the transgene. Retinas exposed to a low titer virus showed no significant inflammatory reaction throughout the observation period, and also maintained the transgene expression over a 1-year period. In the retinas exposed to a high titer virus, however, mononuclear cell infiltration persisted in the subretinal area, and the retina that corresponded to the injected area finally underwent degeneration by around day 90. No retinal neoplastic lesions could be found in any animals over the 1-year period. We thus propose that SIV-mediated stable gene transfer might be useful for ocular gene transfer; however, more attention should be paid to avoiding complications when administering high titer lentivirus to the retina.


Retinitis pigmentosa (RP) is an inherited disease, affecting approximately one in 3500 individuals.1,2 Night blindness and a progressive loss of peripheral visual field are the typical symptoms of RP and a degeneration of photoreceptor cells, such as rods and subsequently cones, also results in a loss of central vision. RP is caused by a mutation in various genes, including rhodpsin, cGMP phosphodiesterase β-subunit (PDE-β), and rds/peripherin, which are expressed in photoreceptor cells.3,4,5 Despite clinical treatments, RP is intractable and still remains a major cause of blindness in adults.

Recently, retinal gene transfer has been suggested as a possible therapeutic strategy for RP. There are basically two concepts for this strategy; one is normal gene supplementation directly to photoreceptor cells, and the other is the gene transfer of secreting neurotrophic factors to the surrounding cells indirectly to prevent photoreceptor cell degeneration. Some investigators have tested several vectors in animal models of RP. In early studies, a recombinant adenovirus was employed for in vivo retinal gene transfer, and showed a transient rescue of photoreceptor cells in rd mice.6,7,8 However, adenoviral vector could evoke a host immune response resulting in a transient gene expression. As a result, it is most likely not suitable as a treatment for human subjects. Furthermore, a transient impairment of the retinal function as assessed by electroretinograms (ERGs) was apparently found, and closely correlated to the degree of histological inflammatory reaction following adenovirus injection.9 ERGs might therefore be an important sensitive assessment technique to detect retinal damage because of vector injection, as well as to identify photoreceptor degeneration.

Several alternatives, including recombinant adeno-associated virus (rAAV) and human immunodeficient virus (HIV)-based lentivirus vectors, are now under investigation,10,11,12,13,14,15,16 and a significant therapeutic effect has been reported in rds mice using rAAV.17 Both of these newcomers seem to be potentially effective in the treatment of retinal diseases, because of their ability to transfer genes to nondividing cells and to sustain transgene expression.15,18 However, possible safety concerns, particularly in the case of lentiviral vectors based on pathogenic viruses including HIV, have delayed their clinical application.

To minimize the potential risks of lentivirus vectors, we recently developed a novel lentivirus vector based on the simian immunodeficiency virus from African green monkeys (SIVagm), which has been reported to be a nonpathogenic virus even in the natural hosts including African green monkeys as well as macaques.19 One possible advantage for safety concerns when using a lentiviral vector from other species is possibly a lower frequency of recombination between the vector sequences and the HIV viral genome.

To obtain preclinical information regarding this new vector in retinal gene transfer, we assessed (1) the in vivo gene transfer efficiency of SIV-based lentivirus vector to the retina of adult rats via the subretinal injection route, (2) the subquantitative evaluation of the transgene expression using two different reporter genes for over 1 year, and (3) the histopathological findings, corresponding to the retinal functions as assessed by ERGs, following the vector injection.


Assessment of the retinal functions via ERGs following the subretinal injection of SIV-NLS-lacZ

To assess the precise correlation among the retinal functions, the transgene expression, and the histological damage of the retina, we first injected two different titers of SIV vector solution, consisting of a low titer (2.5 × 107 TU/ml) and a high titer (2.5 × 108 TU/ml), as well as buffer solution (BSS), into the subretinal space.

We first recorded dark-adapted and light-adapted ERGs at days 3, 10, 30, and 90. As shown in Figure 1, the representative dark- (Figure 1a) and light-adapted (Figure 1b) ERG waveforms at day 3 did not differ among each group. The amplitudes of both the a and b waves, however, decreased at day 3 in all subretinal injection groups. These ERG amplitudes were more greatly depressed in the eyes treated with both titers of vectors than in those treated with BSS, and furthermore the ERG amplitudes were the lowest in the high titer group (Figure 2). Thereafter, these ERG amplitudes gradually recovered in a time-dependent manner. The amplitudes of both the a and b waves at day 30 were not significantly different between the control and both subretinal vector injection groups (Figure 2). These findings may thus indicate that the transient retinal dysfunction followed by subretinal vector injection was partly both due to traumatic damage and to the concentration of the vector solution, and the SIV-mediated gene transfer thus seems to be clinically acceptable regarding the retinal function.

Figure 1

Representative waveforms assessed by ERGs at 3 days after subretinal injection, indicating a dark-adapted response (a) and a light-adapted response (b). The waveform, a and b waves, and implicit time of each component did not differ among the groups (control: untreated control; BSS: 20 μl of buffer injection; SIV-low: injected with 20 μl of 2.5 × 107 TU/ml of SIV-NLS-lacZ; SIV-high: injected with 20 μl of 2.5 × 108 TU/ml of SIV-NLS-lacZ).

Figure 2

Time course alterations of each mean amplitude of dark-adapted a wave (a), dark-adapted b wave (b), and light-adapted 2-Hz b wave (c), assessed by ERGs. At each time point, there are five randomly selected animals in each group (control: untreated control; BSS: 20 μl of buffer injection; SIV-low: injected with 20 μl of 2.5 × 107 TU/ml of SIV-NLS-lacZ; SIV-high: injected with 20 μl of 2.5 × 108 TU/ml of SIV-NLS-lacZ). The mean amplitude±SEM is indicated for each time point. The amplitudes of both the a and b waves decreased at day 3 in each group and thereafter gradually recovered over time. At day 30, the amplitude of ERGs returned to the control level. *P<0.05, **P<0.01 (versus untreated control).

Histological assessment of SIV-mediated lacZ gene transfer

Since the ERG findings were encouraging, we next assessed the SIV-mediated gene transfer by X-gal histochemistry studies using these eyes at various time points (days 7, 14, 30, 60, 100, 120, and 180). The overall histological results are summarized in Table 1.

Table 1 Summary of histological examination for retinal tissue treated with SIV-NLS-lacZ

LacZ gene transfer by a low titer SIV

In the case of a low titer, in situ X-gal staining revealed that a blue area was macroscopically detected at day 7 (Figure 3a), and thereafter maintained a similar blue intensity at 120 and 180 days after injection (Figures 3b and c), which corresponded to the injected area.

Figure 3

Dissecting microscopic view (a–c) and corresponding histopathological findings (d–f) of the rat eye injected with a low titer (2.5 × 107 TU/ml) of SIV-NLS-lacZ at several time points after gene transfer. The eyes are subjected in situ to X-gal histochemistry for 3 h at room temperature. (a–c) Dissecting microscopic views of rat eyes at days 7 (a), 120 (b), and 180 (c). Note the strong and diffuse blue positive staining. Original magnifications; a and b: × 8; c: × 10. (d–f) Histological findings of rat eyes at days 7 (d), 120 (e), and 180 (f), corresponding to (a–c), respectively. Note the frequent and strong blue staining in the RPE up to 180 days (arrows). There was no significant inflammatory reaction and the ONL did not show any significant reduction of the number of layers (bipolar arrows) counterstained with nuclear fast red. Original magnifications; d–f: × 400.

A histological assessment demonstrated that the retinal pigment epithelium (RPE) in the vector-injected area showed frequent X-gal-positive nuclei with neither significant inflammatory infiltrate nor granulation tissue at each time point (Figures 3d-f, arrows). In all retinal tissue specimens, the outer nuclear layer (ONL: Figures 3d-f, bipolar arrows), which mainly contains photoreceptor cells, never showed a significant reduction in the number of layers. In addition, since the other retinal architecture was not affected by vector injection, these results suggest that SIV did not evoke a significant host response to the vector at a low titer. However, only three eyes (total 34 eyes, 8.8%) demonstrated a partial or complete loss of the photoreceptor cell layer, as is frequently seen in the retina exposed to a high titer SIV.

LacZ gene transfer by a high titer SIV

In the case of a high titer, in situ X-gal stain revealed that a blue area was macroscopically detected at day 7 (Figure 4a), and these findings were similar to those of a low titer. On the other hand, the histology of retinal tissue treated with a high titer virus also showed X-gal-positive blue spots in the RPE (Figure 4b, arrowhead), which were circumscribed by ED1-positive mononuclear cell infiltration (Figure 4c) in the subretinal area at day 7. This finding was not observed in the retina using a low titer SIV (Figure 4d). These infiltrating cells were negative for anti-CD2 and control IgG antibodies (data not shown). At day 60, the RPE layer frequently disappeared, and the number of ONL decreased to six to eight layers (Figure 4e, bipolar arrows). At that time, ONL was directly surrounded by granulation tissue containing fibroblasts and chronic inflammatory cells (Figure 4e, arrowheads). At day 120, the retinal tissue was regionally going to be thin and degenerated, and the photoreceptor cells in the area coinciding with the area of subretinal gene transfer had almost completely disappeared (Figure 4f) in the majority of the retina treated with a high titer SIV. At day 180, a gross observation of eye ball stained with X-gal demonstrated a frequent loss of blue spots (Figure 4g). Histopathological findings showed that the retina demonstrated in Figure 4g had completely degenerated in the areas receiving the subretinal injection (Figure 4h).

Figure 4

Dissecting macroscopic view (a, g) and histopathological findings (b, e, f, h) of rat eyes injected with a high titer (2.5 × 108 TU/ml) of SIV-NLS-lacZ at several time points after gene transfer. Immunohistochemical findings of the rat eyes at day 7, using a high titer (c) or a low titer (d) of SIV-NLS-lacZ. The eyes are subjected in situ to X-gal histochemistry for 3 h at room temperature. (a, f) Dissecting microscopic view of the rat eye at days 7 (a) and 180 (f). Note the strong and diffuse blue positive staining at day 7; however, the eye balls that stained for X-gal demonstrated a frequent loss of blue spot original magnifications: × 10. (b, e, f, h) Histopathological findings of rat eyes at days 7 (b), 60 (e), 120 (f), and 180 (h). Note the blue staining in the RPE and mononuclear cell infiltration (arrowheads) at day 7, and thereafter the blue reaction and RPE frequently disappeared at day 60 and later, and the retinal tissue that corresponded to a transduced area completely degenerated. The number of ONLs also decreased in a time-dependent manner (bipolar arrows). Hematoxylin and eosin staining. Original magnifications: × 400. (c, d) Immunohistochemical findings of rat eyes at day 7. Immunohistochemically, the infiltrating mononuclear cells were positive for anti-ED1 antibody (arrowheads), but they were negative for anti-CD2 and control IgG antibodies (data not shown). Counterstained with hematoxylin. Original magnifications: × 400.

One-year follow-up of GFP gene expression

We tried to observe gene expression in the same eyes using GFP as a reporter transgene. Similar to the experiments of lacZ gene transfer, we injected two titers of SIVagm-GFP (2.5 × 107 or 2.5 × 108 TU/ml) into the subretinal space and recorded GFP expression in the retina using a fundus camera at various time points after gene transfer (days 2, 7, 16, 29, 61, 90, 125, 150, 181, 240, 300, and 365). The histological results are summarized in Table 2.

Table 2 Summary of histological examination for retinal tissue treated with SIV-GFP: 1 year after gene transfer

In the case of a low titer, extensive fluorescence of GFP was found, and the extent and frequency was maintained for 1 year (Figures 5a-d). A histological examination revealed that the retinal architecture was well preserved, with neither any significant inflammatory reaction nor photoreceptor cell degeneration (Figures 5e and f), thus indicating the results to be similar to those with lacZ gene transfer. In the case of retinas treated with a high titer vector, however, the extent of GFP expression decreased in a time-dependent manner (Figures 6a-e). A histological analysis for these eyes at day 365 demonstrated that the area receiving gene transfer had undergone degeneration (Figure 6h), while the other part of the retina without vector intervention had been relatively preserved (Figure 6g), thus suggesting that retinal degeneration may be because of gene transfer via SIV.

Figure 5

A 1-year follow-up study for GFP expression in the retinal fundus assessed by fundus camera (a–d) and histopathological findings (e, f) of the eye treated with a low titer SIV. Similar findings were obtained in three of four eyes (see Table 2). (a–d) Retinal GFP expression at days 16 (a), 125 (b), 181 (c), and 365 (d). The area and strength of GFP expression remained almost completely stable for 1 year. (e, f) Longitudinal section of the same eye in (a–d). The section is almost cut in the same part to that of the line in (d). (f) A high-powered view of the squared area of (e). The retinal structure was almost completely preserved. Original magnifications; e: × 50; f: × 400. (e, f) Hematoxylin and eosin staining. ONL (bipolar arrows).

Figure 6

A 1-year follow-up study for GFP expression in the retinal fundus assessed by fundus camera (a–e) and histopathological findings (f–h) of the eye treated with a high titer SIV. Similar findings were obtained in four of five eyes (see Table 2). (a–e) Retinal GFP expression at days 16 (a), 61 (b), 125 (c), 300 (d), and 365 (e). The extent of the area with GFP expression apparently decreased since day 125, and was observed to be sparse at 365 days. (f) Longitudinal section of the same eye in (a–e). The section is almost cut in the same part to that of the line in (e). (g, h) High-powered view of (f). The retina receiving the gene transfer became thin and degenerated; however, the other retina remained almost completely intact. Original magnifications; f: × 50; g, h: × 400. (f–h) Hematoxylin and eosin staining. ONL (bipolar arrows).


In this study, we assessed the long-term effects following reporter gene transfer by an SIVagm-based lentivirus vector on histopathology and ERGs in the adult rat retina. The key observations obtained in this study were that: (1) although a functional disturbance assessed by ERGs was transiently found in a dose-dependent manner following SIV injection, this almost completely recovered by around 1 month after gene transfer, (2) transgene expression via SIV-based lentivirus mainly in RPE in adult animals, and was stable and sustained over a 1-year period, especially when using an appropriate titer of the virus, without any neoplastic lesions, and (3) some retinal tissue treated with SIV, particularly when using a high titer vector solution, demonstrated a sustained inflammatory reaction, the formation of granulation tissue, a loss of transgene expression, and a degeneration of photoreceptor cells associated with breakdown of RPE. The main importance of this report is that it demonstrates a 1-year assessment of safety for the use of lentivirus vector in retinal tissue, and these findings thus imply some useful and practical information regarding the clinical usage of this type of vector.

Regarding the tropism of lentiviruses pseudotyped with VSV-G in retinal tissue, there are discrepant results among the reports. A previous study demonstrated that HIV-based vector could deliver transgenes efficiently to both photoreceptor cells and RPE in newborn rats (age 2–5 days), while the transgene expression was limited around the injection site in young adult rats (age 4 weeks).15 Further, a recent report showed that HIV-based vector targeted both photoreceptor cells and RPE, while transgene expression by HIV-based vector pseudotyped with Mokola was restricted to RPE in adult mice.20 In contrast, we could rarely find an X-gal-positive reaction in the photoreceptor cells, but mainly found it in the RPE at various time points after gene transfer in adult rats using SIV, and similar findings in both adult mice and newborn pups have been reported from other laboratories using HIV- and SIV-based lentivirus.21,22 The exact mechanism of these discrepant tropisms (among infant and aged animals) is unclear. However, one possible explanation may be that the age-related differences in permeability of the vector particles in the retinal tissue. Another possible explanation regarding the different transgene expression patterns may depend on the use of promoters. A previous report demonstrated a more prominent expression in photoreceptor cells in use of rhodpsin promoter, while RPE-dominant expression was found in use of CMV promoter.15 However, CMV promoter may also work well in photoreceptor cells when using AAV,20 thus further studies should be required to clarify this point.

Considering the clinical setting, patients with RP generally recognize the typical symptoms of RP in adulthood and many forms of RP are caused by mutations in photoreceptor-related genes, such as rhodpsin, cGMP PDE-β and rds/peripherin, which are expressed in photoreceptor cells.3,4,5 It may thus be necessary for us to develop strategies to treat RP using gene delivery of secretory molecules such as neurotrophic factors, but not of cytoplasmic proteins in the use of lentiviruses.

The inflammatory response against vectors and/or transgene is an important factor affecting the retinal architecture and functions, as previously demonstrated by us and others in the use of adenovirus for retinal gene transfer.9,23,24 An important finding obtained in this study is that the histological damage of the retina might occur after around days 60–90 when the functional damage assessed by ERGs had almost recovered (Table 1). In the early phase during functional impairment of retina (−30 days), both the transgene expression and the retinal architecture did not show any serious changes, except for a mild decrease in the number of ONLs in the high titer group (Figure 3-6). These results suggest that the functional impairment as assessed by ERGs in early phase (−30 days) may reflect the additive effects of a traumatic detachment of the retina and the amount of vector particles, probably vector-related inflammatory consequences. Regarding the ERG assessment after 90 days in some animals treated with lacZ and GFP vectors, although the numbers were small (n=2 on day 90, n=3 on day 180, and n=2 on day 365), including eyes with severe photoreceptor loss, and all these eyes showed almost normal ERG amplitudes of both a and b waves. The reason of these paradoxical results are unclear, and one possible explanation is that the area of retinal degeneration because of high titer vector injection was so relatively small (approximately 1/6–1/3 of the total retinal area) that ERG did reflect the net retinal function masking the functional disturbance of the damaged area. In a future study, we will assess this aspect using large animals by ‘multifocal ERG’, which can detect the retinal function in each part.

Another significant finding demonstrated in this study is that the dose of the lentivirus vector may be a critical factor for avoiding unexpected deleterious effects because of gene transfer, as shown in Figure 4 and Figure 6. Importantly, the majority of retinal tissue with degeneration was found after the recovery of ERGs (Table 1). The exact reason for the gradual loss of transgene expression, the progressive retinal degeneration, and the sustained inflammatory reaction, seen in retinas treated with a high titer virus, remains unclear. One possible explanation is an unexpected host immune response to the leaky viral proteins from the vector sequence. This hypothesis, however, does not seem to be likely, because the current construct of SIV is a self-inactivating (SIN) vector which lacks any promoter activity in its LTRs,19 and theoretically expresses no virus-related genes except for the transgene driven by an exogenous promoter in the expression cassette. This finding may be supported by a few recent papers, including Doi et al,25 who demonstrated that no retinal damage was seen in the eyes treated with empty HIV-based lentivirus vector. Another hypothesis is a host immune response against the transgene products, including β-galactosidase and GFP, which can evoke CTL and cytotoxic antibodies.26,27,28 Recent reports also suggested that the gradual reduction of GFP expression might be because of a host immune response against GFP itself when using the lentivirus vector.22,25 However, these findings may not be sufficient to conclude these issues, because they used only one reporter gene (GFP). In contrast, we herein demonstrated that a loss of transgene expression was found in a dose-dependent manner and the expression of both GFP and lacZ were not rejected when using low titer viruses. Moreover, although we measured the total IgG levels against the β-galactosidase following subretinal injections of both titers of vector solution, no significant difference was seen between both the high and low viral titers (data not shown). To explain these paradoxical results, extensive investigations, including injection of an empty vector without transgene and a vector devoid of envelop, are thus called for.

Another remaining question is how much vector-related regional degeneration, corresponding to the vector-injected area, affects the whole retinal function. This seems important for us to estimate the patients’ risks for vision. We are presently assessing this aspect, and the results may be obtained within 1 year.

Clearly, major advantage for safety concerns regarding SIVagm may be the possible lower frequency of recombination between SIVagm and HIV. Other possible issues, however, still remain to be overcome. Our SIV-based self-inactivating lentivirus is now in the developmental stage, and a packaging construct contains all accessory genes except for env, the so-called second-generation vector. The SIV vector production system using the packaging construct, eliminating all accessory genes and tat, the so-called ‘third-generation’, should be established before moving to the clinical application stage. Furthermore, one study demonstrated that the experimental inoculation of wild SIVagm to pig-tailed macaques, but not to African green monkeys, resulted in an AIDS-like syndrome,29 thus suggesting that wild SIVagm is not always nonpathogenic in any primate. Further extensive studies assessing safety are thus called for before clinical studies can be performed. We now plan to conduct such studies in the next year using non-human primates with third-generation vectors.

In summary, the current safety assessment in small animals suggests that the SIVagm-based lentivirus vector may be a useful tool for the long-term transgene delivery to retinal tissue and may thus be clinically applicable. More attention should be paid, however, to the unexpected deleterious effects on retinal tissue when using higher concentrations. Although possible limitations remain regarding the use of lentiviruses, this mode of vector still offers a great potential as a treatment modality in the field of retinal gene therapy.

Materials and methods

SIV vectors

Recombinant SIVs were produced as previously described.19 The SIVs encoding the E. coli lacZ gene with Simian Virus 40 large T antigen nuclear localizing signal (SIV-NLS-lacZ) or the green fluorescent protein (SIV-GFP) were propagated. A U3 region in the 3′ and 5′ long terminal repeat (3′ and 5′ LTR) of the SIV was deleted to induce self-inactivation and the SIV vectors were pseudotyped with vesicular stomatitis virus envelope glycoprotein G (VSV-G). The virus titer was determined by a transduction of the human embryonic kidney 293 T cell line as expressed in transducing units/ml (TU/ml), and these viruses were kept at −80°C until just before use. Vector stocks were confirmed to be free from endotoxin, and without extraordinary cytotoxicity by simultaneous transfection testing using 293 T cells and human RPE cells (ARPE-19) obtained from American Type Culture Center (Rockville, MD, USA).

Gene transfer into rat retina

Male Wistar rats (7 weeks old) were used (total n=91; BSS: n=20; SIV-low: n=38; SIV-high: n=33) and maintained in a humane manner, with the proper institutional approval, and in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. All animal experiments were done under the approved protocols and in accordance with the recommendations for the proper care and use of laboratory animals by the Committee for Animals’, Recombinant DNA, and Infectious Pathogens’ Experiments at Kyushu University and according to the Law (No. 105) and Notification (No. 6) of the Japanese Government.

The subretinal injection of each solution was performed as previously described.30 Intraperitoneal injections of pentobarbital (1 mg/kg) were used to induce sufficient anesthesia. For all subretinal injections, we used an operating microscope to monitor related events. First, a 30-gauge needle was inserted into the anterior chamber at the limbus and the anterior chamber fluid was drained off. Next, a 30-gauge needle was inserted into the subretinal space of the peripheral retina via an external trans-scleral transchoroidal approach. A total of 20 μl of SIVagm vector solution (2.5 × 107 or 2.5 108 TU/ml) or balanced salt solution (BSS: 137 mM NaCl, 5.4 mM KCl, 0.44 mM KH2PO4, 0.34 mM Na2HPO4, and 13 mM Tris, pH 7.6) was injected, and any excess solution from the injection site was washed out using phosphate-buffered saline (PBS: 137 mM NaCl, 3 mM KCl, 8 mM Na2HPO4, and 1 mM KH2PO4, pH 7.2). Approximately 5 μl of solution remained in the subretinal space. The appearance of a dome-shaped retinal detachment confirmed the subretinal delivery. Any eyes that sustained prominent surgical trauma, such as retinal or subretinal hemorrhage or bacterial infection, were excluded from this examination.


The ERGs were recorded at days 3, 10, 30, and 90 after both titers of SIV-NLS-lacZ and BSS injection by a masked examiner (YG) without any information regarding the treatments. At each time point, five animals in each group were randomly selected by an independent scientist (YI), and were then transferred to the examination room in a blinded manner. The rats were kept in a dark room at least overnight before the examination. The electroretinographic evaluations were performed by previously described methods.31,32 The rats were anesthetized with an intraperitoneal injection of 15 ml/g body weight of saline solution containing ketamine (1 mg/ml), myoblock (0.4 mg/ml), and urethane (40 mg/ml). Both pupils were dilated with 2.5% phenylephrine, and the animals were placed on a heating pad to maintain the body temperature at a stable level. ERGs were recorded using a coiled stainless-steel wire containing the anesthetized (1% proparacaine HCl) corneal surface through a layer of 1% methylcellulose, and a similar wire was also placed in the leads, respectively. The responses were differentially amplified (band pass; 0.8–1200 Hz), averaged and stored using a PC 9801 computer (NEC, Tokyo, Japan). Xenon strobe flash stimuli (t<1 ms) were presented in a Ganzfeld stimulator (VPA-10; Cadwell, Kennewick, WA, USA), either in the dark or superimposed on an adapting field.

The ERGs were recorded in two sessions. In the first session, the dark-adapted response was obtained using a flash intensity at 1.30 log cd s/m2. The responses to five successive flashes in each rat were averaged. In addition, the 1-min interflash interval was used for this session. The animals were then exposed to a 1.00 log cd s/m2 achromatic-adapting field luminance, and after a period of 20 min during which responses reached a stable amplitude, light-adapted 2 Hz ERG was recorded using flash intensity at 1.30 log cd s/m2. In each rat, the responses to 50 successive flashes were averaged.

X-gal histochemistry

The rats were killed by an overdose of pentobarbital, and the eyes were enucleated and fixed with ice-cooled 2% paraformaldehyde with 0.25% glutaraldehyde in 0.1 M PBS for 10 min. Next, the eyes were washed with 0.1 M PBS containing 0.1% Triton-X and followed by X-gal staining (solution: 5 mM potassium ferrous cyanide, 5 mM ferric cyanide, 2 mM magnesium chloride, 1 mg/ml 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) for 3 h at room temperature. The X-gal-stained tissue was refixed and mounted in paraffin, and 5 μm-thick sections stained with hematoxylin and eosin, or counterstained with nuclear fast red, were examined under a light microscope.


An immunohistochemical analysis was performed to evaluate the inflammatory infiltrating cells. The specimens were incubated overnight at 4°C with one of the primary antibodies: mouse monoclonal anti-rat ED1 IgG antibody (Serotec, Raleigh, NC, USA) and diluted 1:500 for monocyte/macrophage, mouse monoclonal anti-rat CD2 IgG antibody (BD Pharmingen, San Diego, CA, USA) diluted 1:500 for T-lymphocyte and the isotype control, mouse monoclonal anti-rat IgG antibody (BD Pharmingen). Then signals were developed using the avidin-biotinylated peroxidase complex method.

Detection of GFP using fundus camera

Ophthalmoscopy was performed at various time points after gene transfer (days 2, 7, 16, 29, 61, 90, 125, 150, 181, 240, 300, and 365). To detect the GFP in the retina, we used a fundus camera (TRC-50X; Topcon, Tokyo, Japan), which is widely available as a clinical tool for fluorescein angiography (FAG). The rats were anesthetized, and the pupil was dilated with 2.5% phenylephrine (Santen, Osaka, Japan), and the animals were placed on a pad. After we recorded the fundus camera at day 365, the rats were killed and the eyes were enucleated for a histological examination. The eyes were fixed and mounted in paraffin, and 5 μm-thick sections stained with hematoxylin and eosin were examined under light microscopy.

Histological examination

To evaluate the histological changes because of vector injection, we examined the sections of ocular tissue treated with SIV-NLS-lacZ or SIV-GFP under light microscopy. We classified the histological damage into three grades as follows: 0, no decrease in the photoreceptor cell layer; I, a partial loss of photoreceptor cells; II, a complete loss of photoreceptor cells corresponding to the injected area.

Statistical analyses

All values were expressed as the mean±s.e.m. The data were analyzed by one-way ANOVA and, where appropriate, Student's t-test with Scheffe's adjustment for multiple comparisons was used.


  1. 1

    Pagon RA . Retinitis pigmentosa. Surv Ophthalmol 1988; 33: 137–177.

  2. 2

    Wong P . Apoptosis, retinitis pigmentosa, and degeneration. Biochem Cell Biol 1994; 72: 489–498.

  3. 3

    Dryja TP et al. A point mutation of the rhodopsin gene in one form of retinitis pigmentosa. Nature 1990; 343: 364–366.

  4. 4

    Kajiwara K et al. Mutations in the human retinal degeneration slow gene in autosomal dominant retinitis pigmentosa. Nature 1991; 354: 480–483.

  5. 5

    Cotran PR et al. Genetic analysis of patients with retinitis pigmentosa using a cloned cDNA probe for the human gamma subunit of cyclic GMP phosphodiesterase. Exp Eye Res 1991; 53: 557–564.

  6. 6

    Bennett J et al. Photoreceptor cell rescue in retinal degeneration (rd) mice by in vivo gene therapy. Nat Med 1996; 2: 649–654.

  7. 7

    Cayouette M, Gravel C . Adenovirus-mediated gene transfer of ciliary neurotrophic factor can prevent photoreceptor degeneration in the retinal degeneration (rd) mouse. Hum Gene Ther 1997; 8: 423–430.

  8. 8

    Bennett J et al. Adenovirus-mediated delivery of rhodopsin-promoted bcl-2 results in a delay in photoreceptor cell death in the rd/rd mouse. Gene Therapy 1998; 5: 1156–1164.

  9. 9

    Sakamoto T et al. Retinal functional change caused by adenoviral vector-mediated transfection of LacZ gene. Hum Gene Ther 1998; 9: 789–799.

  10. 10

    Ali RR et al. Gene transfer into the mouse retina mediated by an adeno-associated viral vector. Hum Mol Genet 1996; 5: 591–594.

  11. 11

    Flannery JG et al. Efficient photoreceptor-targeted gene expression in vivo by recombinant adeno-associated virus. Proc Natl Acad Sci USA 1997; 94: 6916–6921.

  12. 12

    Bennett J et al. Real-time, noninvasive in vivo assessment of adeno-associated virus-mediated retinal transduction. Invest Ophthalmol Vis Sci 1997; 38: 2857–2863.

  13. 13

    Lewin AS et al. Ribozyme rescue of photoreceptor cells in a transgenic rat model of autosomal dominant retinitis pigmentosa. Nat Med 1998; 4: 967–971.

  14. 14

    Bennett J et al. Stable transgene expression in rod photoreceptors after recombinant adeno-associated virus-mediated gene transfer to monkey retina. Proc Natl Acad Sci USA 1999; 96: 9920–9925.

  15. 15

    Miyoshi H et al. Stable and efficient gene transfer into the retina using an HIV-based lentiviral vector. Proc Natl Acad Sci USA 1997; 94: 10319–10323.

  16. 16

    Takahashi M et al. Rescue from photoreceptor degeneration in the rd mouse by human immunodeficiency virus vector-mediated gene transfer. J Virol 1999; 73: 7812–7816.

  17. 17

    Ali RR et al. Restoration of photoreceptor ultrastructure and function in retinal degeneration slow mice by gene therapy. Nat Genet 2000; 25: 306–310.

  18. 18

    Naldini L et al. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 1996; 272: 263–267.

  19. 19

    Nakajima T et al. Development of novel simian immunodeficiency virus vectors carrying a dual gene expression system. Hum Gene Ther 2000; 11: 1863–1874.

  20. 20

    Auricchio A et al. Exchange of surface proteins impacts on viral vector cellular specificity and transduction characteristics: the retina as a model. Hum Mol Genet 2001; 10: 3075–3081.

  21. 21

    Bainbridge JW et al. In vivo gene transfer to the mouse eye using an HIV-based lentiviral vector; efficient long-term transduction of corneal endothelium and retinal pigment epithelium. Gene Therapy 2001; 8: 1665–1668.

  22. 22

    Duisit G et al. Five recombinant simian immunodeficiency virus pseudotypes lead to exclusive transduction of retinal pigmented epithelium in rat. Mol Ther 2002; 6: 446–454.

  23. 23

    Sakamoto T et al. A vitrectomy improves the transfection efficiency of adenoviral vector-mediated gene transfer to Muller cells. Gene Therapy 1998; 5: 1088–1097.

  24. 24

    Reichel MB et al. Immune responses limit adenovirally mediated gene expression in the adult mouse eye. Gene Therapy 1998; 5: 1038–1046.

  25. 25

    Van Ginkel FW et al. Intratracheal gene delivery with adenoviral vector induces elevated systemic IgG and mucosal IgA antibodies to adenovirus and beta-galactosidase. Hum Gene Ther 1995; 6: 895–903.

  26. 26

    Van Ginkel FW et al. Adenoviral gene delivery elicits distinct pulmonary-associated T helper cell responses to the vector and to its transgene. J Immunol 1997; 159: 685–693.

  27. 27

    Stripecke R et al. Immune response to green fluorescent protein: implications for gene therapy. Gene Therapy 1999; 6: 1305–1312.

  28. 28

    Doi K et al. Lentiviral transduction of green fluorescent protein in retinal epithelium: evidence of rejection. Vision Res 2002; 42: 551–558.

  29. 29

    Johnson PR et al. Molecular clones of SIVsm and SIVagm: experimental infection of macaques and African green monkeys. J Med Primatol 1990; 19: 279–286.

  30. 30

    Ikeda Y et al. Recombinant sendai virus-mediated gene transfer into the retinal tissue of adult rats: efficient gene transfer by brief exposure. Exp Eye Res 2002; 75: 39–48.

  31. 31

    Goto Y et al. Functional abnormalities in transgenic mice expressing a mutant rhodopsin gene. Invest Ophthalmol Vis Sci 1995; 36: 62–71.

  32. 32

    Goto Y . An electrode to record the mouse cornea electroretinogram. Doc Ophthalmol 1996; 91: 147–154.

Download references


We thank H Fujii, Y Hori, and R Hashimoto for assistance with the experiments. Mr Brian Quinn provided language assistance. This work was supported by a Grant of Promotion of Basic Science Research in Medical Frontier of the Organization for Pharmaceutical Safety and Research.

Author information

Correspondence to Y Yonemitsu.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Ikeda, Y., Goto, Y., Yonemitsu, Y. et al. Simian immunodeficiency virus-based lentivirus vector for retinal gene transfer: a preclinical safety study in adult rats. Gene Ther 10, 1161–1169 (2003) doi:10.1038/

Download citation


  • lentivirus
  • simian immunodeficiency virus
  • retina
  • electroretinogram

Further reading